Update README.md

README.md

@@ -13,15 +13,15 @@ should probably proofread and complete it, then remove this comment. -->
 
 # phi-1_5-pl-v_0_1
 
-This model is based on [microsoft/phi-1_5](https://huggingface.co/microsoft/phi-1_5). It is trained from scratch on the 20231201 Polish Wikipedia dump.
+This model is based on [microsoft/phi-1_5](https://huggingface.co/microsoft/phi-1_5). It was trained from scratch on the 20231201 Polish Wikipedia dump.
 
 ## Model description
 
-More information needed
+The model was trained for a context length of 1024 tokens.
 
 ## Intended uses & limitations
 
-More information needed
+The model is intended for research purposes only. It may generate fictitious, incorrect, unethical, or biased texts. At its current state, it is not suitable for production purposes.
 
 ## Training and evaluation data
 
@@ -50,11 +50,13 @@ The following hyperparameters were used during training:
 - lr_scheduler_type: cosine
 - lr_scheduler_warmup_ratio: 0.1
 - num_epochs: 2
+- precision: bf16
 - seed: 42
 
 ### Training results
 
-train_loss: 2.727
+- runtime: 2d 21h 26m 36s
+- train_loss: 2.727
 
 ### Framework versions